Database Design Method for Speech Synthesizers Applied to the Rehabilitation of Disabled People

The speech synthesizer can be used as an electronic rehabilitation device for the speech disabled people. To obtain the highly desirable individual voice, an appropriate speech sample database has to be collected. This paper presents a simple method of determining the voice similarity by analyzing a set of very short sounds. It can be used to shorten the time needed for database collection by selecting a readymade database, most similar to the patient’s voice. The voice similarity was defined as the sum of single elementary similarities of short sound pairs. To find the similarity between voices, a large number of pitch-length sounds were extracted from natural voice signals from different realizations of open vowels 'a' and 'o'. This method is oriented to the use of the microphonemic speech synthesis based on waveform concatenation. This simple and computationally low-cost speech synthesis method can be applied in small portable devices and used for the rehabilitation of speech disabled people.